Backgrounds/Aims

Most gastric cancer cells acquire epithelial-mesenchymal transition (EMT) to promote malignant cells to form metastasis in gastric cancer cases. Recent studies proved the roles of microRNAs on the regulation of EMT through suppressing EMT-related transcription factors. Of those, miR-506 inhibits expression of SNAI2 and PRRX1 and aberrant low expression of miR-506 relates to poor prognosis in ovarian and breast cancers. The aim of this study is to investigate the role of miR-506 on survival and EMT in Japanese gastric cancer (GC) patients.

Methods

144 patients with surgically resected primary GC were included in this study. MicroRNA and mRNA were measured using real-time PCR analysis. Clinicopathological factors including prognosis that were dependent on miR-506 expression were analyzed. In vitro, we established miR-506 overexpressing human GC cell lines (MKN-7 and MKN-45). The proliferation, migration and luciferase assay of 3′UTR of SNAI2 were performed.

Results

(1) The low expression of miR-506 was significantly correlated with poor overall survival in the univariate model (log-rank p = 0.02 ) and multivariate model (hazard ratio = 1.78, p = 0.05). The miR-506 expression was inversely correlated with SNAI2 expression (p = 0.01). The low expression of miR-506 was significantly correlated with histologically poorer differentiation that showed more invasive properties (p = 0.02). (2) In miR-506 overexpressed GC cell lines, miR-506 suppressed the SNAI2 expression by binding to 3′UTR of SNAI2 gene, and resulted in the increased expression of E-cadherin. MiR-506 overexpression significantly suppressed the cell proliferation and the cell migration compared to the miR-control transfected cells.

Conclusions

This study showed that miR-506 directly controlled EMT by regulating SNAI2, and was a independent prognostic factor in Japanese GC patients. Our data indicated that miR-506 could be a candidate of EMT inhibitor in GC patients.

Univariate and multivariate analysis of clinicopathological features for 5-year overall survival 
  Univariate analysis    Multivariate analysis  
Features HR 95%CI P value  HR 95%CI P value 
Age (>70/70≥) 0.626 0.345-1.089 0.099  
Gender (male/female) 1.491 0.844-2.781 0.174  
Histological grade(Well & Moderately/Poorly & others) 1.647 0.959-2.892 0.071     
Depth(T1, 2/T3, 4) 4.957 2.389-12.040 <0.001*  1.932 0.777-5.392 0.162 
Lymph node metastasis (absent/present) 5.363 2.581-13.038 <0.001*  2.937 1.245-7.951 0.012* 
Venous invasion (absent/present) 3.164 1.859-5.420 <0.001*  1.364 0.753-2.500 0.307 
Peritoneum metastasis (absent/present) 4.933 2.635-8.928 <0.001*  2.690 1.391-5.062 0.004* 
Stage I, II/III, IV 5.429 2.948-10.780 <0.001*  
miR-506 expression (low/high) 2.017 1.149-3.713 0.014*  1.775 1.002-3.300 0.049* 
*p<0.05, HR hazard ratio, CI 95% confidence interval        
Univariate and multivariate analysis of clinicopathological features for 5-year overall survival 
  Univariate analysis    Multivariate analysis  
Features HR 95%CI P value  HR 95%CI P value 
Age (>70/70≥) 0.626 0.345-1.089 0.099  
Gender (male/female) 1.491 0.844-2.781 0.174  
Histological grade(Well & Moderately/Poorly & others) 1.647 0.959-2.892 0.071     
Depth(T1, 2/T3, 4) 4.957 2.389-12.040 <0.001*  1.932 0.777-5.392 0.162 
Lymph node metastasis (absent/present) 5.363 2.581-13.038 <0.001*  2.937 1.245-7.951 0.012* 
Venous invasion (absent/present) 3.164 1.859-5.420 <0.001*  1.364 0.753-2.500 0.307 
Peritoneum metastasis (absent/present) 4.933 2.635-8.928 <0.001*  2.690 1.391-5.062 0.004* 
Stage I, II/III, IV 5.429 2.948-10.780 <0.001*  
miR-506 expression (low/high) 2.017 1.149-3.713 0.014*  1.775 1.002-3.300 0.049* 
*p<0.05, HR hazard ratio, CI 95% confidence interval        

Citation Format: Shotaro Sakimura, Junji Kurashige, Keishi Sugimachi, Masami Ueda, Hidenari Hirata, Yoshiaki Shinden, Etsuko Sakimura, Tae Matsumura, Yuki Takano, Ryutaro Uchi, Hiroki Ueo, Hidetoshi Eguchi, Tomoya Sudo, Sumio Hoka, Koshi Mimori. Decreased expression of miR-506 induced epithelial-mesenchymal transition and poor prognosis in gastric cancer patients. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 1478. doi:10.1158/1538-7445.AM2014-1478